⚛️ quantum physics

Quantum-Limited Distance Estimation in Three-Dimensional Optical Superresolution

This paper derives the quantum-limited precision for estimating the three-dimensional distance between two incoherent point sources, demonstrating that sub-Rayleigh information is finite and governed by the point-spread function's second-order displacement-response tensor, which provides a geometric strategy for optimizing resolution through system rotation.

Original authors: Junyan Li, Shengshi Pang

Published 2026-08-06

📖 4 min read🧠 Deep dive

CC BY 4.0

Original authors: Junyan Li, Shengshi Pang

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you are trying to take a picture of two fireflies hovering very close together in the dark. In the world of old-school photography, there's a rule called the "Rayleigh limit" that says if those two lights get too close, they blur into a single, fuzzy blob. For a long time, scientists thought this meant the information about how far apart they really were had simply vanished forever. But then, a new field called "quantum superresolution" came along and whispered a secret: the information isn't actually gone; it's just hiding in the subtle patterns of light waves that our regular cameras can't see. It's like the fireflies are speaking a secret language of ripples that only a very special listener can understand. This paper dives into the three-dimensional version of this mystery, asking a big question: if we have a camera that can hear that secret language, how well can we measure the exact distance between those two fireflies if they are floating in 3D space, and does the shape of our "listening device" matter?

The authors of this paper, Junyan Li and Shengshi Pang, have built a mathematical map to answer exactly that. They show that even when two light sources are practically touching, we can still measure the distance between them with a finite, usable precision, provided we use the right quantum tricks. The key discovery is that the "shape" of the blur—the way the camera spreads out the light—acts like a directional filter. If the blur is perfectly round (like a ball), the distance is hard to pin down no matter how you look at it. But if the blur is stretched out like a rugby ball or a pancake (anisotropic), the distance becomes much easier to measure if you line up your camera just right.

Think of the imaging system as a pair of stretchy, invisible gloves that catch the light. If you squeeze two objects together, the way the gloves deform tells you how far apart the objects are. The paper proves that these gloves have "principal directions"—like the long axis and the short axis of an egg. If you align the two fireflies along the short, tight axis of the glove, the deformation is huge and easy to measure. If you align them along the long, loose axis, the deformation is tiny and hard to spot. The authors found that the "quantum limit" (the absolute best precision nature allows) is determined by a specific mathematical object called a "tensor," which is just a fancy way of describing how the glove stretches in different directions.

Here is the really cool part: you don't always have to move the fireflies to get a better picture. The paper suggests that if you physically rotate your camera (or the "glove"), you can align its tightest stretch direction with the gap between the sources. This is like turning a rugby ball so its pointy ends face the gap you are trying to measure. By doing this, you can squeeze the most information out of the light, potentially making your measurement much sharper. The authors also discovered a neat shortcut: if your "glove" has flat, mirror-like sides (reflection symmetry), you can instantly know which way is the "tight" direction without doing all the heavy math.

To make this concrete, they tested their theory using a "Gaussian" blur, which is just a fancy name for a smooth, bell-shaped cloud of light. They found that for this specific shape, the precision of the distance measurement is directly linked to how "squashed" the cloud is. If the cloud is very flat in one direction, that's the direction where you get the best distance reading. They also showed that if one firefly is much brighter than the other, it gets harder to measure the distance, but the fundamental ability to measure it doesn't disappear completely.

In short, this paper doesn't just say "we can see better"; it gives a geometric rulebook for how to see better. It tells us that the ultimate limit of seeing two close objects isn't just about how powerful your microscope is, but about how you orient your microscope relative to the objects. By understanding the hidden geometry of the light's blur, we can rotate our tools to unlock the best possible view, turning a blurry mess into a precise measurement. This isn't just a theoretical guess; the authors derived these limits mathematically, showing that the information is always there, waiting for us to find the right angle to read it.

Technical Summary: Quantum-Limited Distance Estimation in Three-Dimensional Optical Superresolution

Problem Statement
Conventional optical imaging suffers from the "Rayleigh curse," where the sensitivity to the separation between two incoherent point sources vanishes as their distance decreases below the diffraction limit. While quantum estimation theory has established that this loss of sensitivity in direct intensity measurements does not imply a fundamental loss of information (quantum superresolution), the specific quantum limits for estimating the full physical distance in three-dimensional (3D) imaging systems remain largely unexplored. Specifically, it is unknown how the spatial structure of the point-spread function (PSF) and the intensity imbalance between sources influence the ultimate precision of distance estimation in 3D.

Methodology
The authors model a spatially invariant 3D imaging system containing two incoherent point sources with arbitrary intensity imbalance. The sources are located at positions $\mathbf{s}_1$ and $\mathbf{s}_2$ , separated by a distance $r = |\mathbf{s}_2 - \mathbf{s}_1|$ and a unit displacement vector $\mathbf{u}$ .

Quantum State Formulation: The average state of a detected photon is described by a density operator $\hat{\rho}$ derived from the PSF $\psi(\mathbf{s})$ , which is assumed to be real and inversion-symmetric.
Quantum Fisher Information Matrix (QFIM): The authors calculate the QFIM for the parameters of interest: the centroid position and the displacement vector components. They utilize the fact that the rank of the single-photon density operator is at most two to evaluate the QFIM efficiently.
Multiparameter Estimation: By analyzing the commutator of the symmetric logarithmic derivatives, the authors demonstrate that for real PSFs, the quantum Cramér-Rao bound (QCRB) is attainable, allowing for a precise calculation of the estimation limit for the distance $r$ .
Tensor Analysis: The precision is expressed in terms of a second-order displacement-response tensor, $\mathbf{K}$ , defined by the spatial derivatives of the PSF. The authors analyze the eigensystem of $\mathbf{K}$ and its relationship to the PSF's reflection symmetries.
Gaussian Example: A general 3D Gaussian PSF is used as a specific case study to illustrate the theoretical framework, linking the response tensor directly to the inverse spatial covariance matrix.

Key Contributions and Results

Finite Precision in the Sub-Rayleigh Regime: The study derives the quantum-limited precision for estimating the full distance $r$ between two sources. It proves that even as $r \to 0$ (the sub-Rayleigh regime), the quantum information remains finite, confirming the absence of the Rayleigh curse at the quantum limit for 3D distance estimation.
Governance by the Response Tensor: The estimation precision is governed by the second-order displacement-response tensor $\mathbf{K}$ of the PSF. The precision $H_r$ in the limit $r \to 0$ is given by:
$H_r = \frac{N_{\text{tot}}(1 - \epsilon^2)}{\mathbf{u}^\top \mathbf{K}^{-1} \mathbf{u}}$
where $N_{\text{tot}}$ is the total photon number and $\epsilon$ is the intensity imbalance.
Geometric Optimization via Orientation: The precision depends on the relative orientation $\mathbf{u}$ $u$ between the sources and the principal axes of the PSF.
- The optimal precision is achieved when the displacement vector $\mathbf{u}$ aligns with the eigenvector of $\mathbf{K}$ corresponding to its largest eigenvalue.
- The worst precision occurs when $\mathbf{u}$ aligns with the eigenvector corresponding to the smallest eigenvalue.
- The maximum achievable precision gain is the ratio of the largest to the smallest eigenvalues of $\mathbf{K}$ .
System Rotation Strategy: The paper proposes a practical strategy for superresolution optimization: rather than rotating the sources (which may be physically constrained), one can physically rotate the anisotropic imaging system. Rotating the system is mathematically equivalent to rotating the relative orientation $\mathbf{u}$ , allowing the alignment of the source displacement with the PSF's optimal principal response direction.
Symmetry Identification: The authors show that reflection symmetries of the PSF directly identify the principal axes of the response tensor $\mathbf{K}$ . If the PSF possesses reflection planes, the normal vectors to these planes are eigenvectors of $\mathbf{K}$ , simplifying the identification of optimal orientations without full tensor diagonalization.
Gaussian PSF Connection: For a general 3D Gaussian PSF, the response tensor $\mathbf{K}$ is proportional to the inverse of the spatial covariance matrix ( $\mathbf{K} \propto \Sigma^{-1}$ ). This establishes a direct link between quantum-limited precision and the geometry of the Gaussian distribution. The optimal orientation aligns the source displacement with the minor axis of the Gaussian covariance ellipsoid (the direction of smallest spatial spread).

Significance
The paper establishes a general geometric framework for 3D quantum superresolution. It demonstrates that the ultimate precision of distance estimation is not merely a function of source separation but is intrinsically linked to the spatial structure and orientation of the imaging system's PSF. By revealing that the eigensystem of the displacement-response tensor dictates the optimal and worst-case scenarios, the work provides a direct strategy for improving resolution: engineering or physically rotating anisotropic imaging systems to align their principal response directions with the source displacement. This geometric perspective connects the physical properties of the imaging system directly to the fundamental limits of estimation performance.

Technical Summary: Quantum-Limited Distance Estimation in Three-Dimensional Optical Superresolution

More like this